International Journal of Mathematics and Computer Applications Research (IJMCAR) ISSN 2249-6955 Vol.

2, Issue 2 June 2012 1-13 TJPRC Pvt. Ltd.,

A SET OF QUALITATIVE MEASURES TO IMPROVE QUALITY OF WEBSITE CONTENT

G. SREEDHAR Assistant Professor, Department of Computer Science Rashtriya Sanskrit Vidyapeetha (Deemed University) Tirupati, India.

ABSTRACT
Over the last few years there has been a remarkable increase in use of the World Wide Web (WWW) for a wide and variety of purposes. There was also a fast growth in its applications. This led the Internet users to realize the importance and the benefits gained from a globally interconnected hypermedia system. The quality model must be able to assess the quality of each and every aspect of the website and it should cover the process of all web engineering activities. A set of guidelines are considered to establish the procedure for Quality Assessment Process (QAP). A set of guidelines are considered to establish the procedure for Quality Assessment Process (QAP). The qualitative measures are identified in in QAP are divided into Text

Formatting Measures (TFM), Graphics Element Measures (GEM), Page Formatting Measures (PFM),Link Formatting Measures (LFM),Page Performance Measures (PPM)

KEYWORDS: Website Content, QAP, TFM, PFM, GEM, LFM, PPM INTRODUCTION
Over the last few years there has been a remarkable increase in use of the World Wide Web (WWW) for a wide and variety of purposes. There was also a fast growth in its applications. This led the Internet users to realize the importance and the benefits gained from a globally interconnected hypermedia system. On the other hand it causes a larger number of useless, meaningless and badly designed websites on the Internet world causing unwanted additional traffic; this is all because of an unorganized non-planned websites development processes. Due to the unceasing growth of web sites and applications, developers and evaluators have interesting challenges not only from the development but also from the quality assurance point of view.

RELATED WORK
As we know, the quality assurance was and is one of the challenging processes in software engineering as well as for the web engineering, as a new discipline. Although there exists many design guidelines, and metrics for the evaluation of web sites and applications, most of them lack a well-defined specification framework and even worse a strategy for consultation and reuse. Some initial efforts have been recently made to classify metrics for some entity type as

G. Sreedhar

for example metrics for software products. Particularly, in last few years a set of web site metrics were defined and specified based on the data collection point of view. The quality model must be able to assess the quality of each and every aspect of the website and it should cover the process of all web engineering activities. A set of guidelines are evolved to build a qualitative model of a website. A guideline consists of a design and evaluation principle to be observed to get and to

guarantee a usable user interface [1]. Guidelines can be found in many different formats with contents varying both in quality and level of detail, ranging from ill-structured common sense statements to formalized rules ready for automatic guidelines checking . Certain rules are

validated by experimental results provided by user tests, experiments in laboratory or other techniques. Guidelines can be classified (Figure 1) by type ranging from the most general to the most specific: principles, guidelines and recommendations. Principles are general objectives guiding conceptual User Interface (UI) decisions. They reflect the knowledge around human perception, learning and behavior and are generally expressed in generic terms like Use images and metaphors consistent with real world so that they can be applied for a wide range of cases. Guidelines are based on principles specific to a particular design domain. For example, a web design rule can stipulate to use a consistent look and a visual language inside the site. Some guidelines have to be interpreted more and altered to reflect the needs of a particular organization or a design case. Recommendations determine conceptual decisions specific to a particular domain of application and should reflect the needs and the terminology of a given organization. They are unambiguous statements so that no place for interpretation is left. Recommendations include ergonomic algorithms, user interface patterns and design rules. Design rules are functional and operational requirements specifying the design of a particular interface, e.g. Every web page needs an informative title

Principles

Guidelines

Recommendations

Figure 1.Types of guidelines and sources Kwaresmi [2] developed a framework to define a Guideline Definition Language (GDL) to investigate quality evaluation procedure. The GDL expresses guideline information in a sufficiently rich manner so that evaluation engine can perform GDL-compliant guideline.

A Set of Qualitative Measures to Improve Quality of Website Content

U(p)= f kwaresmi (Web_page,UESi, j )= EXEC(ECi, j {INST_UESi, j }) = {"Respected" |"Violated" |"PartiallyRespected"} ... (1)

Where UESi,j be the set of evaluation sets associated to the guideline i in the source j and that will be used for the evaluation of the evaluated web page. ECi,j be the set of evaluation conditions associated to UESi,j. INST_UESi,j is the set of captured instances of UESi,j in the evaluated page. In practice, the f (Web _ page,UESi , j ) executes each ECi,j condition and then it combines the results to have the overall result for the guideline i. We say that a web page satisfies a guideline Gi,j, if the execution of all ECi,j on all the INST_UESi,j is true. Using the above evaluation parameters allows us to define a kind of quality model to balance the evaluation result. In the accessibility field, Bobby [3] defined a set of accessible evaluation tools. All these tools are based on accessibility guidelines. It does this through automatic checks as well as manual checks. It also analyzes web pages for compatibility with various browsers (equation 2). Accessibility tools use a binary model to evaluate the accessibility of web pages.
guidelines

Accessibility errors =

i=1

ai xi ...

(2)

where ai is 0 when guideline is violated and 1 when guideline is not violated and xi is a guideline. An easy way to comply with the conference paper formatting requirements is to use this document as a template and simply type your text into it.

METHODOLOGY TO OPTIMIZE QUALITY OF WEBSITE CONTENT

A set of guidelines are considered to establish the procedure for Quality Assessment Process (QAP). The World Wide Web Consortium (W3C) [4] is an open source organization and it defines various web standards for designing a website. The W3C is led by web inventor Tim Berners-Lee and CEO. The standards defined by W3C are considered as guidelines and these guidelines help in assessing the quality of website content in presenting the web content. The guidelines are as summarized as follows. Guideline 1: Provide a text equivalent for every non-text element. This includes images, graphical representations of text, image map regions, animations, applets and programmatic objects, frames, scripts, spaces, audio and video files. Guideline 2: Do not rely on color scheme only. The content of web page must match with foreground and background color. Also provide sufficient contrast to the content for visibility.

G. Sreedhar

Guideline 3: Use markup and style sheets instead of images to convey information. Style sheets controls the layout and presentation of the web page and decreases the download time of the web page. Guideline 4: Clearly mention the text information of web page with natural language. Specify the expansion of each abbreviation or acronym in the document. Guideline 5: Use tables properly in the web document. For data tables, clearly specify row and column headers and number of rows and columns exactly. Guideline 6: Ensure that web pages featuring new technologies transform gracefully. When dynamic contents are updated, ensure that content is changed. Ensure that pages are available and meaningful when scripts, applets or other programmatic objects are not supported by the browsers. If this is not possible, provide equivalent information as alternative in the web page. Guideline 7: Ensure user control of time sensitive content changes. Until user agents provide the ability to stop the refresh, do not create periodically auto-refreshing pages. Guideline 8: Ensure direct accessibility of embedded user interfaces. Make programmatic elements such as scripts and applets directly accessible or compatible with assistive technologies. Guideline 9: Design for device-independence. Ensure that any element that has its own interface can be operated in a device-independent manner. Guideline 10: Provide context orientation information. Title each frame to facilitate frame identification and navigation. Divide large blocks of information into more manageable groups wherever appropriate. Guideline 11: Provide clear navigation mechanisms. Clearly identify the target of each link. Provide information about the general layout of a site such as site map or table of contents. Guideline 12: Ensure that documents are clear and simple. Create a style of presentation that is consistent across pages. A set of qualitative measures are identified to satisfy the quality factors for optimizing website content. These measures derived from the Web page errors that are generated using the W3C Validation Service. This process uses the standard web tool W3C HTML Validator to validate and identify the number of different errors according syntax errors of HTML tags, properties of web page and standards mentioned by various organizations such as W3C. Most pages on the World Wide Web are written in computer languages (such as HTML) that allow Web authors to structure text, add multimedia content, and specify what appearance or style, the result should have. As for every language, these have their own grammar, vocabulary and syntax, and every document written with these computer languages are supposed to follow these rules. Markup languages are defined in technical specifications, which generally include a formal grammar. The tool compares HTML document to the defined syntax of HTML and reports any

A Set of Qualitative Measures to Improve Quality of Website Content

discrepancies. The outputs of the Markup Validator are a list of error messages and their interpretation W3C HTML validator helps to ensure that documents are free of potential problems that can result in unexpected output when users view the bad documents with different browsers. A screenshot of W3C HTML Validator is shown in figure 2.

Figure 2: W3C HTML Validator

The errors related to website content cause incorrect display of some components of Web pages. These errors include:

i. ii. iii. iv.

Table Tag Errors (TTE): All the sub tags in table tag should be properly used in the web page design. Errors in table tag cause for display problems of web page. Body Tag Errors (BTE): Body Tag Errors cause the errors in displaying the contents of the web page. Image Tag Errors (ITE): Image Tag Errors cause for errors in downloading the image in a website. Head Tag Errors (HTE): Head Tag Errors cause for errors in displaying heading and title of the web page.

v.

Font Tag Errors (FoTE): Font Tag Errors cause the errors in textual display of the web page.

vi.

Script Tag Errors (STE): Script Tag Errors cause the errors in programming at client side scripting.

vii.

Style Tag Errors (StTE): Style Tag Errors cause errors in dynamic display features of the web page.

6 viii.

G. Sreedhar

Form Tag E rrors (FmTE): Form Tag Errors cause errors in input and output display of the script programming in a web page.

ix.

Link Tag Errors (LTE): Link Tag Errors cause errors in linking various web components.

The web content errors are occurred due to non-standards of web site design. The developer must be attentive in using HTML tags so that appropriate tags should be used in web design process. All the tags and their attributes properly set and closed accordingly. This will reduce the problem of web page display and avoids the problem in downloading of the web page. These errors are further used to identify qualitative measures for website content. The qualitative measures are further divided into various types. They are 1. 2. 3. 4. 5. Text Formatting Measures (TFM) Graphics Element Measures (GEM) Page Formatting Measures (PFM) Link Formatting Measures (LFM) Page Performance Measures (PPM)

The website contents should be optimized in such a way that all the elements of the web pages of website should display correctly without any loss of presentation. The contents of website must be presented in such a way that the user has to get 100% satisfaction in viewing all web pages of the website so that complete content of website visible and all pages are accessible. A) Quality Assurance using Text Formatting Measure: Text Presentation is an important issue in display of the web content. These issues should be properly handled in presenting 100% correct text presentation. Several literature sources provide guidance about appearance of text on the web page. To format the text on web page the developer should consider following properties in text presentation. Fonts must be chosen among the most readable ones. Font size must be defined as relative size Use fonts designed for computer screens rather than fonts designed for print In a single page, the number of different fonts must be limited. When using different fonts and font sizes, they should have some specific meaning (e.g. notes, links, and navigation location). Avoid italicizing and underlining text

A Set of Qualitative Measures to Improve Quality of Website Content

These properties can be detected and measured by parsing both the text and CSS. The above properties are defined as attributes in various HTML tags. These tags include <HEAD>, <BODY> and <FONT> tags. The errors in attributes of tag errors cause incorrect or semi correct display of text on web page. The algorithm QAPM identifies the errors related to text presentation. Thus the Head Tag Errors (HTE), Font Tag Errors (FoTE) and Body Tag Errors (BTE) identify the problems in the text presentation of web page. Thus Text Formatting Measures are to be evaluated for correct text presentation. Text formatting measures are evaluated using the metric

TFM and it is defined in equation 3

.... (3)

n n n TFM = BTEi + FoTEi + HTEi n i =1 i =1 i =1

B) Quality Assurance using Link Formatting Measures: The link presentation is important aspect in organization of web pages. TThe links in a website may be internal or external links. There are differences among the internal and external links. While internal links must all be valid and links pointing to external domains are out of control of the webmaster, but can be checked. Link topology is an often neglected aspect. Some sites are just trees of nodes, with links from a node pointing to children and to ancestors. Some others have a much more complex link topology, with many horizontal or traverse links. To format the links on the web page perform following functions. Use moderate levels of breadth with minimal in the information architecture. Minimize depth Avoid broken links Use corresponding text links Redundant links may cause confusion and avoid them Effective navigation requires small pages, few clicks between pages and strong scent The algorithm QAPM identifies Link Tag Errors (LTE) and broken links related to above properties. The Link Formatting Measures are evaluated from LTE and broken links of the website and Link Formatting Measures are very much necessary for efficient usage of hyperlinks in the website. The LFM is measured using the metric defined in equation 4.

n LFM = BL + SCE + LTE n *100 1 i=

i i i

... (4)

C) Quality Assurance using Page Formatting Measures: The page layout is probably the principal characteristic perceived by the user. Layout must be clean, and the whole content should be well structured. A page layout is designed using tables, <div> tag or <frame> tag. Layout

G. Sreedhar

must adaptable to different devices. This implies that page must avoid making reference to specific device settings, like screen resolution or fixed size page components. An automated analysis of CSS usage and coding can supply information about the layout and the adoption of an organization wide standard. The algorithm QAPM generates table tag errors (TTE), frame tag errors (FTE), style tag errors (StTE), font tag errors (FoTE), frame tag usage errors and document type declaration errors if any attribute of tag element deviate the properties of page formatting. The Page Formatting Measures assess the following features of website. Use browser-safe colors Use no more than 6 discriminable colors Use 256 (8-bit) color palettes Avoid using black backgrounds Use high contrast between background and text Keep line lengths to 40-60 characters Keep text between 9 to 15 words per line Avoid using framesets Text should cover no more than 25-30% of the screen Greater text density facilitates page scanning Page Formatting Measures (PFM) evaluates the page layout quality of various web pages of the website. The metric is defined in equation 5.

n PFM =TTEi +FTEi +StTEi +FTUEi +DTDEi n .... (5) i=1

D) Quality Assurance using Graphic Formatting Measures: The graphic presentation is the important issue in presenting pictorial and multimedia components. To format Graphics on the web page perform following properties should be followed. Avoid using graphical text links Use corresponding text links instead of graphical links Avoid using animation unless it is appropriate Proper contrast between foreground image and background (color or image) The image tag error (ITE), body tag errors (BTE) and image load errors related to image identifies the errors in display of images and hence Graphic Element Measures to be evaluated. Graphics Element Measures developed for assessing the following features of web interfaces. The

A Set of Qualitative Measures to Improve Quality of Website Content

Graphic Element Measures (GEM) evaluate the quality of multimedia presentation using the defined in equation (6).
n G E M = IT E i + IL E i i =1 n ... (6)

E) Quality Assurance using Page Performance Measures: correctness is a merely technical aspect, which can be easily checked. In many cases inconsistent behaviour with different

browsers can be originated by lack of conformance to the published grammars (HTML, XHTML) and the default actions taken by the browsers themselves. Correctness is easily checked as an internal quality factor. The important aspects to consider in some environments are the professionalism and effectiveness of the web site that could be measured through how many different platforms are supported and it supports adaptivity and adaptability for a personalization [5], [6]. The form tag errors (FmTE), script tag errors (STE) and title tag with no keyword errors identify the need of page performance measure. The page performance measures developed to answer the following questions related to the website. To increase the performance of the web page following guidelines are considered. Avoid gratuitous use of technology Minimize the use of video Avoid using sound files Effective navigation requires small pages Avoid using Click Here for link text] Page Performance Measures (PPM) evaluates the correctness of the web pages display of the website. The PPM is measured using the metric defined in equation 7.
n PPM = STEi + FmTEi + TTKEi n i =1 ... (7)

Having defined various qualitative measures at this part of the research work, it is tried to evolve 10-point scale. Thus 10-point scale is a metric towards defining quality of web content. The 10-point quality metric mainly depends on the qualitative measures In this connection it is interpreted the 10-point scale indicates such that 0 always represent poorer side and 10 always represent the best side of quality aspect. The metrics for various qualitative measures are formulated using empirical evaluation. Here 10-point scale of web content depends on the value computed using measures and its level competence based o performance. The values of 10-point scale of a metric suggest important remarks for the web designer in order to improve the quality of website. The values are grouped into various categories after thorough analysis of experimental results. The description of 10-point scale is shown in table 1.

10

G. Sreedhar

Table 1: 10-Point Scale Description

10-Point Scale Values 0 to 4 5 to 6 7 8 9 10 Remarks Very Poor Poor Need Improvement Need minor changes Good Very Good

The 10-point scale values for various qualitative measures is defined using equations 8 to 12.

10 if TFM 2.5 9 if TFM 5 8 if TFM 7.5 7 if TFM 10 6 if TFM 12.5 TFM10POINTMETRIC = 5 if TFM 15 4 if TFM 17.5 3 if TFM 20 2 if TFM 22.5 1 if TFM 25 0 otherwise .... (8)

A Set of Qualitative Measures to Improve Quality of Website Content

11

10 if LFM 2.5 9 if LFM 5 8 if LFM 7.5 7 if LFM 10 6 if LFM 12.5 LFM 10 POINTMETRIC = 5 if LFM 15 4 if LFM 17.5 3 if LFM 20 2 if LFM 22.5 1 if LFM 25 0 othrwise .... (9)

10 if PFM 2 9 if PFM 4 8 if PFM 6 7 if PFM 8 6 if PFM 10 PFM10POINTMETRIC = 5 if PFM 12 4 if PFM 14 3 if PFM 16 2 if PFM 18 1 if PFM 20 0 otherwise ..... (10)

12

G. Sreedhar

10 if GEM 2.5 9 if GEM 5 8 if GEM 7.5 7 if GEM 10 6 if GEM 12.5 GEM10POINTMETRIC = 5 if GEM 15 4 if GEM 17.5 3 if GEM 20 2 if GEM 22.5 1 if GEM 25 0 otherwise .... (11)

10 if PPM 2.5 9 if PPM 5 8 if PPM 7.5 7 if PPM 10 6 if PPM 12.5 PPM10POINTMETRIC = 5 if PPM 15 4 if PPM 17.5 3 if PPM 20 2 if PPM 22.5 1 if PPM 25 0 otherwise .... (12)

A Set of Qualitative Measures to Improve Quality of Website Content

13

CONCLUSIONS
The main theme of the research paper is to provide some qualitative measure for clear and quality web content of website. It is observed that website must be informative and all contents of the website must be accommodated in page layout according to standard guidelines. An attempt is made to enhance the quality of website content and layout so that web designer shall follow the quality of content in designing a web site.

REFERENCES
[1]. Dreyfus P (1998), Principles of usability, http://www.devedge.netscape.com/viewsource/arcive/editior98_3_23.html [2]. Beirekdar A, Vanderdonckt J, Noirhomme-fraiture M: A Framework and a Language for Usability Automatic Evaluation of Web sites by static Analysis of HTML Source Code: Kolski Ch, Vanderdonckt J (eds): Proceedings of 4th International Conference on Computer-Aided Design of User Interfaces CADUI (2002). [3]. [4]. [5]. [6]. Bobby, http://webxact.watchfire.com/ www.w3c.org Etnoteam S.P.A, marzo 2000, 1995-96, available at http://ww.etnoteam.it/webquality M.Y. Ivory, An Empirical foundation for Automated Web Interface Evaluation, doctoral dissertations, University of California, Berkeley, Computer Science Department, 2001.